[Beowulf] GPU-HMMer for interested people
landman at scalableinformatics.com
Wed Dec 17 07:51:00 PST 2008
Olli-Pekka Lehto wrote:
> Joe Landman wrote:
>> Hi folks
>> GPU-HMMer (part of the MPI-HMMer effort) has just been
>> announced/released at http://www.mpihmmer.org
>> MPI-HMMer has itself been improved with parallel-IO and better
>> scalability features. JP has measured some large number (about 180x)
>> over single cores on a cluster for the MPI run.
> Hi Joe,
> Looks quite promising. Here are results from a simple real-world test case:
> GPU: Dual GTX280, each with 1GB RAM
> CPU: Single Intel Core2 quad Q9550 2.83GHz
> hmmsearch 4 threads sorted: 274.49s
> hmmsearch 4 threads unsorted: 254.23s
> cuda_hmmsearch unsorted 407.85s
> cuda_hmmsearch sorted: 62.69s
> cuda_hmmsearch sorted 2 simultaneous runs: 78.23s 80.79s
> -Running hmmsort to sort the sequence database is critical to obtain
> reasonable performance from cuda_hmmsearch. However, the regular
> hmmsearch is slightly slower with the sorted database.
> -Running two simultaneous runs assigned to different GPUs on a dual-GPU
> quad-core system yields some performance penalty, but is still quite
> -I used the parameters THREADSIZE=320 BLOCKSIZE=64. I'm not completely
> sure if these are the optimum values for GTX280. Any better suggestions?
I'd suggest subscribing/posting to the mpihmmer list
(http://lists.scalableinformatics.com/mailman/listinfo/mpihmmer) so we
don't clutter beowulf with this.
JP could likely tell you more about the appropriate sizing for GTX280.
We have a pair of GTX260's in the lab that some runs were done on.
Could you describe your run a bit more (sequence queries, hmm size,
database and size)? On the mpihmmer list though ...
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web : http://www.scalableinformatics.com
phone: +1 734 786 8423 x121
fax : +1 866 888 3112
cell : +1 734 612 4615
More information about the Beowulf